Efficient pole-search algorithm for dynamic polarizability

Toward alternative excited-state calculation for large systems

    Research output: Contribution to journalArticle

    2 Citations (Scopus)

    Abstract

    This study presents an efficient algorithm to search for the poles of dynamic polarizability to obtain excited states of large systems with nonlocal excitation nature. The present algorithm adopts a homogeneous search with a constant frequency interval and a bisection search to achieve high accuracy. Furthermore, the subtraction process of the information about the detected poles from the total dynamic polarizability is used to extract the undetected pole contributions. Numerical assessments confirmed the accuracy and efficiency of the present algorithm in obtaining the excitation energies and oscillator strengths of all dipole-allowed excited states. A combination of the present pole-search algorithm and divide-and-conquer-based dynamic polarizability calculations was found to be promising to treat nonlocal excitations of large systems.

    Original languageEnglish
    Pages (from-to)7-14
    Number of pages8
    JournalJournal of Computational Chemistry
    Volume38
    Issue number1
    DOIs
    Publication statusPublished - 2017 Jan 5

    Fingerprint

    Excited States
    Excited states
    Search Algorithm
    Pole
    Poles
    Excitation
    Alternatives
    Bisection
    Excitation energy
    Divide and conquer
    Subtraction
    Dipole
    High Accuracy
    Efficient Algorithms
    Interval
    Energy

    Keywords

    • dynamic polarizability
    • excited state
    • linear scaling
    • time-dependent density functional theory

    ASJC Scopus subject areas

    • Chemistry(all)
    • Computational Mathematics

    Cite this

    @article{6a7f6f0dcfc04e1095dcd016741c7065,
    title = "Efficient pole-search algorithm for dynamic polarizability: Toward alternative excited-state calculation for large systems",
    abstract = "This study presents an efficient algorithm to search for the poles of dynamic polarizability to obtain excited states of large systems with nonlocal excitation nature. The present algorithm adopts a homogeneous search with a constant frequency interval and a bisection search to achieve high accuracy. Furthermore, the subtraction process of the information about the detected poles from the total dynamic polarizability is used to extract the undetected pole contributions. Numerical assessments confirmed the accuracy and efficiency of the present algorithm in obtaining the excitation energies and oscillator strengths of all dipole-allowed excited states. A combination of the present pole-search algorithm and divide-and-conquer-based dynamic polarizability calculations was found to be promising to treat nonlocal excitations of large systems.",
    keywords = "dynamic polarizability, excited state, linear scaling, time-dependent density functional theory",
    author = "Hiromi Nakai and Takeshi Yoshikawa and Yutaro Nonaka",
    year = "2017",
    month = "1",
    day = "5",
    doi = "10.1002/jcc.24507",
    language = "English",
    volume = "38",
    pages = "7--14",
    journal = "Journal of Computational Chemistry",
    issn = "0192-8651",
    publisher = "John Wiley and Sons Inc.",
    number = "1",

    }

    TY - JOUR

    T1 - Efficient pole-search algorithm for dynamic polarizability

    T2 - Toward alternative excited-state calculation for large systems

    AU - Nakai, Hiromi

    AU - Yoshikawa, Takeshi

    AU - Nonaka, Yutaro

    PY - 2017/1/5

    Y1 - 2017/1/5

    N2 - This study presents an efficient algorithm to search for the poles of dynamic polarizability to obtain excited states of large systems with nonlocal excitation nature. The present algorithm adopts a homogeneous search with a constant frequency interval and a bisection search to achieve high accuracy. Furthermore, the subtraction process of the information about the detected poles from the total dynamic polarizability is used to extract the undetected pole contributions. Numerical assessments confirmed the accuracy and efficiency of the present algorithm in obtaining the excitation energies and oscillator strengths of all dipole-allowed excited states. A combination of the present pole-search algorithm and divide-and-conquer-based dynamic polarizability calculations was found to be promising to treat nonlocal excitations of large systems.

    AB - This study presents an efficient algorithm to search for the poles of dynamic polarizability to obtain excited states of large systems with nonlocal excitation nature. The present algorithm adopts a homogeneous search with a constant frequency interval and a bisection search to achieve high accuracy. Furthermore, the subtraction process of the information about the detected poles from the total dynamic polarizability is used to extract the undetected pole contributions. Numerical assessments confirmed the accuracy and efficiency of the present algorithm in obtaining the excitation energies and oscillator strengths of all dipole-allowed excited states. A combination of the present pole-search algorithm and divide-and-conquer-based dynamic polarizability calculations was found to be promising to treat nonlocal excitations of large systems.

    KW - dynamic polarizability

    KW - excited state

    KW - linear scaling

    KW - time-dependent density functional theory

    UR - http://www.scopus.com/inward/record.url?scp=85027918739&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85027918739&partnerID=8YFLogxK

    U2 - 10.1002/jcc.24507

    DO - 10.1002/jcc.24507

    M3 - Article

    VL - 38

    SP - 7

    EP - 14

    JO - Journal of Computational Chemistry

    JF - Journal of Computational Chemistry

    SN - 0192-8651

    IS - 1

    ER -